The classification problem is typically addressed using machine learning techniques. However, when there is an imbalance among classes in the data, these models often tend to overpredict the majority classes. As a result, the model exhibits low accuracy for some classes while achieving high accuracy for others. This challenge is particularly evident in lung cancer datasets, which contain 283 positive cases compared to only 38 negative cases. To mitigate this class imbalance, this study employs the Synthetic Minority Oversampling Technique (SMOTE), a well-known data preprocessing method. By comparing the performance of various machine learning models before and after applying this balancing technique, we identified Logistic Regression as the top-performing model, achieving an accuracy of 96.03% and a precision of 96.36%. Furthermore, after fine-tuning this model, we obtained an accuracy of 98.15%, an AUC of 99.92%, and a perfect precision of 100%.